Patent Application
20170039196
The present application claims the priority benefit of the filing date of Indian Provisional Application No.: 4104/CHE/2015, filed Aug. 6, 2015, entitled “COMPUTING SYSTEM TO EVALUATE SENTIMENT OF ENTITIES BY PROCESSING DATA”.
Embodiments of the invention generally relate to data processing, and more particularly to computing system to evaluate sentiment of entities by processing data.
Enterprises provide procurement applications where buyers procure goods and services from supplier organizations. Procurement involves managing buying decisions based on various factors associated with the supplier organizations and buyer organizations. Similarly, in auctions, goods and services are transacted based on the bids offered by various suppliers or buyers depending on the context, where the auctions may be real time auctions or non-real-time auctions. Both the procurement scenario and auction scenario rely on static data associated with the suppliers and buyers to manage buying decisions. However, buying decisions made based on the static data are not accurate, since they do not provide real-time information, and behavior of the organizations and users representing the organization at the time of decision making. Real-time information and behavior of the involved organizations and users representing the organizations enable informed decision making.
The embodiments are illustrated by way of examples and not by way of limitation in the figures of the accompanying drawings in which like references indicate similar elements. The embodiments, together with its advantages, may be best understood from the following detailed description taken in conjunction with the accompanying drawings.
Embodiments of techniques of computing system to evaluate sentiment of entities by processing data are described herein. In the following description, numerous specific details are set forth to provide a thorough understanding of the embodiments. One skilled in the relevant art will recognize, however, that the embodiments can be practiced without one or more of the specific details, or with other methods, components, materials, etc. In other instances, well-known structures, materials, or operations are not shown or described in detail.
Reference throughout this specification to “one embodiment”, “this embodiment” and similar phrases, means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one of the one or more embodiments. Thus, the appearances of these phrases in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Procurement applications in procurement network enable managing financial data and transactions. The procurement applications set a standard process for procurement of goods and services. Suppliers can register in the procurement application by creating a supplier profile, and providing supplier associated data, answers to questionnaire, etc. Users representing the suppliers can also be registered in the procurement application. These users represent the suppliers during auctions. Similarly, buyer organizations and users representing buyer organizations can also be registered in the procurement application.
When a contract is generated in a procurement application, a past bidding practice of the buyers can have an influence in the generation of the current contract. Similarly, the type of contracts accepted by the suppliers in the past can have an influence in current contract. The contract is influenced based on a behavior or pattern of the buyers and the suppliers. Buyers often depend on the associations they make, and what they have learned about a specific supplier, which can influence buyer decision making. The behavior can be studied by sentiment analysis. Sentiment refers to subjective impressions such as attitude, emotions, opinions, etc. Sentiment analysis is the process of using natural language processing techniques, statistics or machine learning algorithms or methods to extract, identify, or characterize the sentiment content.
In sentiment analysis, subjective information is extracted from several sources from the internet, that provides insights about opinions on suppliers such as from reviews, financial results, rating on service, related recent news, tweets, comments, social networking site conversations, advertisements, etc. The subjective information from these sources is typically in an unstructured form referred to as unstructured data. Unstructured data captured from the sources of web are parsed and processed using natural language processing, text analysis, etc. Sentiment calculation is performed on the processed data to compute/evaluate sentiment value associated with the suppliers and buyers. Sentiment analysis provides valuable insight to the suppliers and buyers in making informed decisions.
The communication between the frontend application system 125 and the server system 130 may be in the form of internal application programming interface (API) calls. An API is a set of routine, protocols and tools that defines functionalities of the enterprise application independent of implementation of the enterprise application. The API calls between the frontend application system 125 and the server system are internal to the enterprise and are referred to as internal API calls. Similarly, the API calls between the partner system 120 and the server system 130 are external to the enterprise and are referred to as external API calls. The API calls from the partner system 120 can be served as pay per request or pay per specific number of requests by the server system 130.
The buyer system 105 and the supplier system 110 have sentiment view modules 135a and 135b respectively, and the server system 130 has a sentiment analysis module 140. Sentiment view module 135a, sentiment view module 135b and sentiment analysis module 140 may be a combination of software program logic and a graphical user interface associated with the procurement application. Various other product systems such as product system A 145, product system N 150 and external feed system 155, interact with the server system 130. Product system A 145 and product system N 150 may be specialized enterprise systems with hardware capable of hosting and executing specialized enterprise software applications. External feed system 155 may be a specialized system with hardware capable of hosting and executing social networking applications. Partner system 120 may be a third-party system with hardware capable of hosting and executing third-party enterprise software applications. The buyer system 105 and the supplier system 110 may receive data from a buyer user and a supplier user. The buyer system 105 and the supplier system 110 may be client systems, where the client systems can be computing devices such as mobile phones, desktop or portable computers, tablet computers, etc.
When a supplier is registered in the procurement application by creating a supplier profile, and providing supplier associated data, answers for questionnaire, etc., data associated with the supplier is sent to the server system 130 via frontend application system 125. Based on the data sent by the supplier, a sentiment value for the supplier is calculated in the sentiment analysis module 140 in the server system 130. The computed/evaluated sentiment value for supplier can be displayed in the sentiment view module 135b in the supplier system 110. Similarly, individual sentiment values can be calculated for supplier users, suppliers, buyers and buyer users respectively in the sentiment analysis module 140. Supplier users may be users representing suppliers and buyer users may be users representing buyers. Calculated sentiment values can be stored in the database system 160. Database system 160 may be an in-memory database that primarily relies on main memory such as RAM (Random access memory) for processing and data storage.
Product system A 145 and product system N 150 may represent any enterprise product that enables collection of data about suppliers or buyers from various other data bases associated with the enterprise product, or various other web sources. The gathered data are sent to the server system 130, and the sentiment analysis module 140 in the server system 130 analyzes the data to calculate sentiment values for the suppliers or buyers. External feed system 155 enables collection of data feeds from tweets, social networking sites or any other source external to the enterprise. Calculated sentiment values are stored in the database system 160.
Based on the sentiment of the extracted ‘keyword’ a relative numeric value is associated with the extracted ‘keyword’. If the extracted ‘keyword’ indicates a positive sentiment, a positive numeric value of sentiment indicator is associated from a scale. If the extracted ‘keyword’ indicates a neutral sentiment, a mid numeric value of sentiment indicator is associated from the scale. If the extracted ‘keyword’ indicates a negative sentiment, a negative numeric value of sentiment indicator is associated from the scale. For a ‘keyword’ ‘excellent’ a sentiment indicator of ‘+9’ can be associated, for a ‘keyword’ ‘poor future’ a sentiment indicator of ‘−8’ can be associated, for a ‘keyword’ ‘average growth’ a sentiment indicator of ‘0’ can be associated, etc. The sum of the individual sentiment indicators gives us a total sentiment indicator for ‘supplier A’ in the ‘general supplier information’ section 205. Similarly, a total sentiment indicator for ‘supplier A’ in the ‘questionnaire’ section 220 is calculated. Sum of the total sentiment indicator for ‘supplier A’ in the ‘general supplier information’ section 205, and the total sentiment indicator for ‘supplier A’ in the ‘questionnaire’ section 220, gives an overall sentiment indicator/sentiment indicator for ‘supplier A’.
Consider an example, where supplier organization ‘supplier A’ and ‘supplier user A’ representing the supplier organization are registered in a procurement application by creating a supplier profile. Similarly, another supplier organization ‘supplier B’ and ‘supplier user B’ representing the other supplier organization are registered in the procurement application by creating another supplier profile. Buyer organization ‘buyer A’ and ‘buyer user A’ representing the buyer are registered in the procurement application by creating a buyer profile. When the buyer organization ‘buyer A’ intends to procure ‘product A’, a request is raised in the procurement application. Various suppliers registered in the procurement application bid their respective price quotation for the indicated quantity of ‘product A’. Sentiment analysis/evaluation of the supplier and buyer entities is explained below in
Subjective information associated with the ‘supplier A’ 302 is dynamically extracted from several sources of web such as data feeds, tweets, comments, recent news, etc., other products, and used in calculation of sentiment value. The subjective factors are external to the supplier organization and they are referred to as ‘external factors’. The ‘external factors’ can be dynamically changed in real-time by adding new factors, and deleting or modifying existing factors. ‘Internal factors’ and ‘external factors’ may be user-defined. For a factor ‘review’ 324, a weight of ‘0.4’ 326 is associated with a source ‘blog A’ 328. Based on the sentiment of keywords such as ‘excellent’ and ‘good’ extracted from source ‘blog A’ 328, a sentiment indicator is calculated as ‘1’ 330. The product of weight ‘0.4’ 326 and the sentiment indicator ‘1’ 330 is the weighted sentiment indicator ‘0.4’ 332 for ‘supplier A’ 302 for the factor ‘review’ 324. For a factor ‘financial result’ 334, a weight of ‘1’ 336 is associated with a source ‘NSE’ 338, and a sentiment indicator is calculated as ‘1’ 340. The product of weight ‘1’ 336 and the sentiment indicator ‘1’ 340 is the weighted sentiment indicator ‘1’ 342 for ‘supplier A’ 302 corresponding to the factor ‘financial result’ 334.
For a factor ‘recent news’ 344, a weight of ‘1’ 346 is associated with a source ‘BBC’ 348, and a sentiment indicator is calculated as ‘1’ 350. The product of weight ‘1’ 346 and the sentiment indicator ‘1’ 350 is the weighted sentiment indicator ‘1’ 352 for the ‘supplier A’ 302 corresponding to the factor ‘recent news’ 344. For a factor ‘tweets’ 354, a weight of ‘0.35’ 356 is associated with a source ‘tweet source A’ 358, and a sentiment indicator is calculated as ‘0.75’ 360. The product of weight ‘0.35’ 356 and the sentiment indicator is ‘0.75’ 360 is the weighted sentiment indicator ‘0.26’ 362 for ‘supplier A’ 302 corresponding to the factor ‘tweets’ 354. Sum of weighted sentiment indicators 314, 322, 332, 342, 352 and 362 is the weighted sentiment value ‘3.66’ 364 of ‘supplier A’ 302. The weighted sentiment value can be displayed in the sentiment view module in the supplier system.
In a similar way, the weighted value of factors is computed/evaluated for entity ‘supplier B’ 366 registered with the procurement application. A sentiment indicator ‘0.35’ 368 is calculated for ‘supplier B’ 366 for the factor ‘general supplier information’ 304. The product of weight ‘1’ 310 and the sentiment indicator ‘0.35’ 368 is the weighted sentiment indicator ‘0.35’ 370 for ‘supplier B’ 366 corresponding to the factor ‘general supplier information 304. A sentiment indicator ‘0.35’ 372 is calculated for ‘supplier B’ 366 for the factor ‘questionnaire’ 306. The product of weight ‘1’ 318 and the sentiment indicator ‘0.35’ 372 is the weighted sentiment indicator ‘0.35’ 374 for ‘supplier B’ 366 corresponding to the factor ‘questionnaire’ 306. For a factor ‘review’ 324, a sentiment indicator is calculated as ‘0.1’ 376. The product of weight ‘0.4’ 326 and the sentiment indicator ‘0.1’ 376 is the weighted sentiment indicator ‘0.04’ 378 for ‘supplier B’ 366 corresponding to the factor ‘review’ 324.
For a factor ‘financial results’ 334, a sentiment indicator is calculated as ‘0.5’ 380. The product of weight ‘1’ 336 and the sentiment indicator ‘0.5’ 380 is the weighted sentiment indicator ‘0.5’ 382 corresponding to the factor ‘financial results’ 334. For a factor ‘recent news’ 344, a sentiment indicator is calculated as ‘0.5’ 384. The product of weight ‘1’ 346 and the sentiment indicator ‘0.5’ 384 is the weighted sentiment indicator ‘0.5’ 386 for ‘supplier B’ 366 corresponding to the factor ‘recent news’ 344. For a factor ‘tweets’ 354, a sentiment indicator is calculated as ‘0.75’ 388. The product of weight ‘0.35’ 356 and the sentiment indicator ‘0.75’ 388 is the weighted value ‘0.26’ 390 for ‘supplier B’ 366 corresponding to the factor ‘tweets’ 354. Sum of weighted sentiment indicators 370, 374, 378, 382, 386 and 390 is the weighted sentiment value ‘1.74’ 392 of ‘supplier B’ 366. The weighted sentiment value can be displayed in the sentiment view module in the supplier system.
The calculation of weighted sentiment value of an entity may be represented in an equation as shown below:
The weighted sentiment value can be mapped to a sentiment representation such as free form text, a graphical representation, etc. A mapping between weighted sentiment value and a sentiment representation can be user-defined and may be changed dynamically based on the context, and stored in a database system. For example, a weighted sum of sentiment value ranging between >=4 and <=6 is mapped to a free form text “Strong market presence and highly recommended”, a weighted sentiment value ranging between >=2 and <4 is mapped to a free form text “average market presence and recommended”, and a weighted sentiment value ranging between >=0 and <2 is mapped to a free form text “below average market presence and not recommended”. The weighted sentiment value ‘3.66’ 364 of ‘supplier A’ 302 is mapped to a sentiment data “average market presence and recommended”. The weighted sentiment value ‘1.74’ 392 of ‘supplier B’ 366 is mapped to a sentiment data “below average market presence and not recommended”. The mapped values are displayed in the sentiment view module of “buyer system”. Buyer may be able to decide that ‘supplier A’ 364 has a strong market presence and is recommended over ‘supplier B’ 366 based on the sentiment evaluation. In one embodiment, if an API request including a parameter ‘supplier=supplier A’ is received from a partner system, the API request is authenticated and validated to determine if the API request is a paid request. When the API request is validated, a response to the API request is provided as ‘average market presence and recommended’. An API request may include any number of parameters.
FIG. 4 is a table illustrating calculation of weighted sentiment values 400 in a procurement application, according to one embodiment. In one embodiment, for entities ‘supplier user A’ and ‘supplier user B’ registered with the procurement application, sentiment evaluation is performed. Various factors associated with the supplier users are profile information, designation, familiarity with number of auctions, number of successful bids, relevant experience, etc. The factors and the sources associated with the factors can be dynamically changed in real-time. Information associated with these factors, mined and extracted from various source enable calculation of sentiment values for the supplier users. A weight is associated with a source depending on the authenticity of the source, importance associated with the source, type of the source, etc. Consider an entity ‘supplier user A’ 402, for a factor ‘successful auctions in current organization’ 404, a weight of ‘1’ 406 is associated with a source ‘corporate DB’ 408. ‘Supplier user A’ 402 was successful in 40 auctions out of 100 auctions in the current organization. Accordingly, a sentiment indicator is calculated as 40/100=‘0.4’ 410. The product of weight ‘1’ 406 and the sentiment indicator ‘0.4’ 410 is the weighted sentiment indicator ‘0.4’ 412 of ‘supplier user A’ 402 for the factor ‘successful auctions in current organization’ 404.
For the factor ‘successful auctions in prior employment’ 414, a weight of ‘0.9’ 416 is associated with a source ‘auction DB’ 418. ‘Supplier user A’ 402 was successful in 150 auctions out of 500 auctions in the prior employment. Accordingly, a sentiment indicator is calculated as 150/500=‘0.3’ 420. The product of weight ‘0.9’ 416 and the sentiment indicator ‘0.3’ 420 is the weighted sentiment indicator ‘0.27’ 422 corresponding to the factor ‘successful auctions in prior employment’ 414. For the factor ‘work experience’ 424, a weight of ‘0.75’ 426 is associated with a source ‘other DB’ 428. ‘Supplier user A’ 402 has an experience of 12 years and accordingly sentiment indicator is ‘0.75’ 430. The product of weight ‘0.75’ 426 and the sentiment indicator ‘0.75’ 430 is the weighted sentiment indicator ‘0.56’ 432 of ‘supplier user A’ 402 for the factor ‘work experience’ 424. Sum of weighted sentiment indicators 412, 422 and 432 is the weighted sentiment value ‘1.23’ 434 of ‘supplier user A’ 402.
Similarly, for the factor ‘successful auctions in current organization’ 404, ‘supplier user B’ 436 was successful in 65 auctions out of 100 auctions in the current organization. Accordingly, a sentiment indicator is calculated as 65/100=‘0.65’ 438. The product of weight ‘1’ 406 and the sentiment indicator ‘0.65’ 438 is the weighted sentiment indicator ‘0.65’ 440 for ‘supplier user B’ 436 corresponding to the factor ‘successful auctions in current organization’ 404. For the factor ‘successful auctions in prior employment’ 414, ‘supplier user B’ 436 was successful in 295 auctions out of 350 auctions in the prior employment. Accordingly, a sentiment indicator is calculated as 295/350=‘0.84’ 442. The product of weight ‘0.9’ 416 and the sentiment indicator ‘0.84’ 442 is the weighted sentiment indicator ‘0.76’ 444 for ‘supplier user B’ 436 corresponding to the factor ‘successful auctions in prior employment’ 414. For the factor ‘work experience’ 424, ‘supplier user B’ 436 has an experience of 14 years, and accordingly a sentiment indicator is calculated as ‘0.8’ 446. The product of weight ‘0.75’ 426 and the sentiment indicator value ‘0.8’ 446 is the weighted sentiment indicator ‘0.6’ 448 corresponding to the factor ‘work experience’ 424. Sum of weighted sentiment indicators 440, 444 and 448 is the weighted sentiment value ‘2.01’ 450 of ‘supplier user B’ 436.
The weighted sentiment values can be mapped to a sentiment representation such as free form text, a graphical representation, etc. For example, a weighted sentiment value ranging between >=1 and <=2 is mapped to a free form text “average bidder”, and a weighted sentiment value ranging between >2 and <=3 is mapped to a free form text “aggressive bidder”. The weighted sentiment value of ‘supplier user A’ ‘1.23’ 434 is mapped to a sentiment representation “average bidder”. The weighted sentiment value of ‘supplier user B’‘ 2.01’ 450 is mapped to a sentiment representation “aggressive bidder”. The mapped values are displayed in the sentiment view module of “buyer system”. The buyer can make a decision that ‘supplier user B’ 436 is an aggressive bidder, in comparison to the ‘supplier user A’ 402.
Some embodiments may include the above-described methods being written as one or more software components. These components, and the functionality associated with each, may be used by client, server, distributed, or peer computer systems. These components may be written in a computer language corresponding to one or more programming languages such as, functional, declarative, procedural, object-oriented, lower level languages and the like. They may be linked to other components via various application programming interfaces and then compiled into one complete application for a server or a client. Alternatively, the components maybe implemented in server and client applications. Further, these components may be linked together via various distributed programming protocols. Some example embodiments may include remote procedure calls being used to implement one or more of these components across a distributed programming environment. For example, a logic level may reside on a first computer system that is remotely located from a second computer system containing an interface level (e.g., a graphical user interface). These first and second computer systems can be configured in a server-client, peer-to-peer, or some other configuration. The clients can vary in complexity from mobile and handheld devices, to thin clients and on to thick clients or even other servers.
The above-illustrated software components are tangibly stored on a computer readable storage medium as instructions. The term “computer readable storage medium” should be taken to include a single medium or multiple media that stores one or more sets of instructions. The term “computer readable storage medium” should be taken to include any physical article that is capable of undergoing a set of physical changes to physically store, encode, or otherwise carry a set of instructions for execution by a computer system which causes the computer system to perform any of the methods or process steps described, represented, or illustrated herein. A computer readable storage medium may be a non-transitory computer readable storage medium. Examples of a non-transitory computer readable storage media include, but are not limited to: magnetic media, such as hard disks, floppy disks, and magnetic tape; optical media such as CD-ROMs, DVDs and holographic devices; magneto-optical media; and hardware devices that are specially configured to store and execute, such as application-specific integrated circuits (“ASICs”), programmable logic devices (“PLDs”) and ROM and RAM devices. Examples of computer readable instructions include machine code, such as produced by a compiler, and files containing higher-level code that are executed by a computer using an interpreter. For example, an embodiment may be implemented using Java, C++, or other object-oriented programming language and development tools. Another embodiment may be implemented in hard-wired circuitry in place of, or in combination with machine readable software instructions.
A data source is an information resource. Data sources include sources of data that enable data storage and retrieval. Data sources may include databases, such as, relational, transactional, hierarchical, multi-dimensional (e.g., OLAP), object oriented databases, and the like. Further data sources include tabular data (e.g., spreadsheets, delimited text files), data tagged with a markup language (e.g., XML data), transactional data, unstructured data (e.g., text files, screen scrapings), hierarchical data (e.g., data in a file system, XML data), files, a plurality of reports, and any other data source accessible through an established protocol, such as, Open Data Base Connectivity (ODBC), produced by an underlying software system (e.g., ERP system), and the like. Data sources may also include a data source where the data is not tangibly stored or otherwise ephemeral such as data streams, broadcast data, and the like. These data sources can include associated data foundations, semantic layers, management systems, security systems and so on.
In the above description, numerous specific details are set forth to provide a thorough understanding of embodiments. One skilled in the relevant art will recognize, however that the embodiments can be practiced without one or more of the specific details or with other methods, components, techniques, etc. In other instances, well-known operations or structures are not shown or described in detail.
Although the processes illustrated and described herein include series of steps, it will be appreciated that the different embodiments are not limited by the illustrated ordering of steps, as some steps may occur in different orders, some concurrently with other steps apart from that shown and described herein. In addition, not all illustrated steps may be required to implement a methodology in accordance with the one or more embodiments. Moreover, it will be appreciated that the processes may be implemented in association with the apparatus and systems illustrated and described herein as well as in association with other systems not illustrated.
The above descriptions and illustrations of embodiments, including what is described in the Abstract, is not intended to be exhaustive or to limit the one or more embodiments to the precise forms disclosed. While specific embodiments and examples are described herein for illustrative purposes, various equivalent modifications are possible within the scope, as those skilled in the relevant art will recognize. These modifications can be made in light of the above detailed description.
| Number | Date | Country | Kind |
|---|---|---|---|
| 4104/CHE/2015 | Aug 2015 | IN | national |
